Introduction
In my recent blog posts, I have been writing about Total Productive Maintenance (TPM) and its impact on manufacturing, facility availability, product costs, customer satisfaction, and many other aspects of how the facilities we manage affect the end-user or customer.
In this post, I want to take some time to define four types of maintenance: reactive, preventive, planned, and predictive. I will discuss when each type should be used, the associated costs, and the downsides of applying each type of maintenance plan.
Examining the Four Types of Maintenance
Reactive Maintenance
Reactive maintenance is the most common type of maintenance, but it is also the most expensive and disruptive. It is also the least effective type of maintenance, as it does not prevent failures.
Reactive maintenance is typically performed when an asset fails and often involves making emergency repairs. This can be costly and disruptive, as it may require shutting down operations and paying for overtime labor.
Reactive maintenance is often used for assets that are not critical to operations or have a low cost of failure. However, it is essential to note that even minor asset failures can lead to significant downtime and costs if not addressed promptly.
Planned Maintenance
Planned maintenance is performed on a schedule, regardless of whether or not the asset is showing any signs of failure. This type of maintenance is less expensive and disruptive than reactive maintenance, but it is not very effective at preventing failures.
Planned maintenance is typically performed on assets that are critical to operations or have a high cost of failure. It is also used for assets that have predictable failure modes.
Examples of planned maintenance tasks include:
•    Changing the oil in a car
•    Lubricating moving parts on a machine
•    Inspecting a roof for leaks
•    Cleaning, inspecting and replacing air filters
•    Testing electrical systems
Planned maintenance can be scheduled in any combination of daily, weekly, monthly, quarterly, semi-annual or annual, depending on the type of asset and its operating environment.
Predictive Maintenance
Predictive maintenance uses data analysis to predict when an asset will fail. This allows maintenance before the asset fails, preventing downtime and unexpected costs. Predictive maintenance is the most effective type of maintenance, but it is also the most expensive to implement.
Predictive maintenance is often used for critical assets with predictable failure modes, such as manufacturing equipment and aircraft engines. It can also be used for assets that are expensive to repair or replace.
Examples of predictive maintenance techniques include:
•    Vibration monitoring
•    Oil analysis
•    Thermal imaging
•    Acoustic emission monitoring
 Predictive maintenance systems typically collect data from sensors attached to the asset. The data is then analyzed to identify trends and patterns indicating a potential failure.
 When a potential failure is detected, a maintenance alert is generated. This allows maintenance to be scheduled before the asset fails.
Reliability-Centered Maintenance (RCM)
RCM is a process that combines elements of all three types of maintenance. RCM involves identifying the failure modes of assets and developing maintenance strategies to mitigate those failure modes. RCM is a complex process, but it can be very effective at reducing maintenance costs and improving asset reliability.
RCM is typically used for critical assets with complex failure modes. It can also be used for expensive assets to repair or replace.
The RCM process typically involves the following steps:
1.    Identify the assets to be included in the RCM program.
2.    Identify the failure modes of each asset.
3.    Analyze the failure modes to determine the consequences of failure, the causes of failure, and the controls that can be implemented to prevent or mitigate failures.
4.    Develop maintenance strategies for each failure mode.
5.    Implement the maintenance strategies and monitor the results.

Cost Impact of Each Type of Maintenance
The maintenance cost varies depending on the type of maintenance being performed and the asset being maintained. However, in general, reactive maintenance is the most expensive type of maintenance, followed by planned maintenance, predictive maintenance, and RCM.
The overarching question is why should you care about the maintenance, and how do you balance maintenance costs with the costs of goods sold?  As you might recall from our article last week, there is a point where the two costs reach an inflection point, and spending more on maintenance will increase your costs of goods sold.

Remember that maintenance is crucial because it helps to:
•    Reduce downtime and unexpected costs
•    Improve the reliability of assets
•    Extend the life of assets
•    Improve safety and compliance
By choosing the right type of maintenance for each asset, businesses can reduce costs, improve reliability, and extend the life of their assets.
Here are some examples of how different types of maintenance can be used to improve business operations:
A manufacturing plant might use predictive maintenance to monitor its production line equipment for signs of failure. This would allow the plant to schedule maintenance before the equipment fails, preventing unplanned downtime and ensuring the production line runs smoothly.
A hospital might use RCM to develop maintenance strategies for its medical equipment. This would help reduce the maintenance cost and improve the equipment's reliability, which is critical to patient care.
A transportation company might use planned maintenance to schedule regular inspections and maintenance of its fleet of vehicles. This would help prevent unplanned breakdowns and ensure the fleet is safe and reliable.
Maintenance is an essential part of any business. By choosing the right type of maintenance for each asset, businesses can reduce costs, improve reliability, and extend the life of their assets.
The relative cost differences in time and money between each type of maintenance strategy can be challenging to quantify, as they vary depending on several factors, including the specific assets being maintained, the operating environment, and the size and complexity of the organization. However, some general estimates can be made. According to a study by the Reliability Engineering Center at the University of Maryland, these costs can be generalized as detailed below.
Reactive Maintenance
Reactive maintenance is the most expensive type of maintenance, costing organizations from 2% to 5% of the total value of their assets each year. It can also lead to significant downtime, ranging from 10% to 20% of the time. This is because it involves repairing or replacing assets after they have failed, which can lead to significant downtime and unexpected costs.
Preventive Maintenance
Planned maintenance is less expensive than reactive maintenance but still more expensive than predictive maintenance and RCM. It costs organizations from 1% to 3% of the total value of their assets each year and can reduce downtime to an average of 5%. This is because preventive/planned maintenance involves performing maintenance tasks on a schedule, regardless of whether or not the asset is showing any signs of failure.
Predictive Maintenance
Predictive maintenance is the most cost-effective over the long term, costing organizations from 0.5% to 1% of the total value of their assets each year and reducing downtime to an average of 2%. This is because it uses data analysis to predict when an asset will likely fail, allowing maintenance to be performed before it fails. This can help to prevent downtime and unexpected costs.
Reliability-Centered Maintenance (RCM)
RCM is the most expensive type of maintenance to implement, but it can also be the most cost-effective over the long term. RCM involves identifying the failure modes of assets and developing maintenance strategies to mitigate those failure modes. This can help prevent failures, extend the life of assets, and reduce the overall maintenance cost. It costs organizations from 2% to 3% of the total value of their assets in the first year of implementation. Still, it can reduce maintenance costs to an average of 1% of the total value of assets over the long term.

Moving Forward
In addition to the maintenance cost, businesses should also consider the cost of downtime when choosing a maintenance strategy. Downtime can lead to lost revenue, productivity losses, and increased costs. The cost of downtime will vary depending on the industry and the specific assets involved.
For example, a manufacturing plant may lose millions of dollars in revenue for every hour of downtime. A hospital may face significant costs for patient care if its medical equipment fails. A transportation company may lose revenue and incur additional costs if its vehicles break down.
Companies need to take the time to look at their maintenance and establish where equipment and asset reliability ranked on a scale from “run-to-fail” up to “cannot go offline.”  This, combined with assessing equipment redundancy, informs the rating of equipment criticality.  Think of it this way: when you have a fully redundant cooling system in Phoenix, AZ, during the summer, having an A/C unit go down is not a super critical issue.  However, if your 83-year-old mother’s single A/C unit goes down, you have a vital issue. It is prudent to have regular bi-annual maintenance on residential A/C units in Phoenix.
In the same way, each organization needs to take the time to evaluate its assets to determine the criticality of the equipment based on the redundancy and impact if the equipment goes down.  Based on this evaluation, a maintenance plan should be developed that details the maintenance that should be done, what resources should be used to complete the maintenance, and the estimated cost to complete the maintenance.
Conclusion
Businesses should carefully consider the cost of maintenance and downtime when choosing a maintenance strategy. For critical assets with a high cost of failure, predictive maintenance or RCM may be the best option, even though these strategies are initially more expensive to implement. For non-critical assets such as drinking fountains, planned maintenance on some components and run-to-fail on others is the correct maintenance plan.